[TensorFlow 2.0] combat three major projects - image classification

Learning immediately: https://edu.csdn.net/course/play/26956/347465?utm_source=blogtoedu

tensorflow generally used procedure:

Import Data -> Define model -> compilation model -> model training -> Save model -> model predictions

Simple classification model

import tensorflow as tf

inputs=tf.keras.Input(shape=[32,32,3])
'''卷积模块'''
x=tf.keras.layers.Conv2D(10,kernel_size=[3,3],strides=[1,1],padding='SAME',activation='relu',name='conv_1')(inputs)
x=tf.keras.layers.AveragePooling2D(pool_size=[2,2],strides=[2,2])(x)
x=tf.keras.layers.BatchNormalization()(x)
'''展平、接入全连接层'''
x=tf.keras.layers.Flatten()(x)
x=tf.keras.layers.Dense(512,activation='relu')(x)
x=tf.keras.layers.Dense(10,activation='softmax')(x)#输出置信度

'''模型实例化'''
model=tf.keras.Model(inputs=inputs,outputs=x)

model.summary()

MobileNet V1

1. Lightweight convolution neural network
2. Fewer parameters, a smaller amount of calculation, but has a decent performance
3. Space separable convolution